Title: Impact of Glucose Control on Hospital Mortality in the Cardiac Intensive Care Unit
Introduction:
Hyperglycemia in the intensive care unit (ICU) is common, even amongst non-diabetics, and may be due to a number of factors including the release of stress hormones, use of exogenous glucocorticoids and catecholamines, and release of mediators in sepsis. SCCM recommends maintaining blood glucose (BG) levels between 150-180mg/dL in medical and surgical ICU patients. However, little is known about the impact of glucose control in the cardiac intensive care unit (CICU) population.
Methods:
This was a retrospective cohort analysis of patients greater than 18 years of age admitted to the University of Michigan CICU (12/2016-12/2020) with at least one BG lab during admission. The study was approved by the University of Michigan Institutional Review Board. The primary outcome was hospital mortality.
Results:
A total of 3217 patients were included. The mean age was 63.6±14.9 years though patients in the cohort with diabetes were significantly older than those patients without diabetes (65.2±12.5 vs 61.9±16.5; p< 0.01). The majority (64.7%) of patients were male and Caucasian (81.2%). There were significant differences in hospital mortality between blood glucose quartiles for those with diabetes (Q1 11.9%, Q2 13.1%, Q3 22.4%, Q4 21.4%; p< 0.01). There was higher mortality with increasing quartiles in those without diabetes (Q1 6.4%, Q2 8.2%, Q3 13.7%, Q4 24.5%; p< 0.01). Significant predictors for hospital mortality on multivariate analyses for patients with diabetes included age, Elixhauser comorbidity score, mechanical ventilation, length of stay, and hypoglycemic event. Significant predictors for hospital mortality on multivariate analyses for patients without diabetes included age, Elixhauser comorbidity score, mechanical ventilation, admission diagnosis, average BG, length of stay, hypoglycemic event and BG >180.
Conclusions:
In this retrospective cohort of 3217 patients admitted to the CICU, hyperglycemia was associated with increased mortality. The highest mortality in patients with and without diabetes was seen in the highest quartile of average blood glucose. Given differences in mortality trends by quartiles in our study between patients with and without diabetes, future studies should be conducted to determine the optimal glucose ranges for critically ill patients admitted to the CICU.